Method for sharing multi-source perception data based on consensus mechanism

Abstract

The application relates to the technical field of data sharing, and discloses a multi-source sensing data sharing method based on a consensus mechanism, which comprises the following steps: classifying power nodes according to a clustering algorithm, establishing a plurality of subnets based on the classification results, and selecting an agent node for each subnet from the power nodes included in the subnet according to a right and interest agent mechanism; information transmission of data in the same subnet is realized by means of a network topology structure; and data of different subnets are connected and transmitted by means of a consensus mechanism. The embodiment of the application greatly improves the sensing data sharing efficiency and the safety of data transmission based on a power wireless sensor network system, realizes efficient sharing and storage of multi-source sensing data by power wireless sensor network edge computing communication nodes in scenarios such as transformer substations and converter stations, and supports state analysis and fault diagnosis of power equipment by edge applications.

Description

Technical Field

[0001] This invention relates to the field of data sharing technology, and specifically to a multi-source sensing data sharing method based on a consensus mechanism. Background Technology

[0002] With the continuous development of power systems and the increase in electricity demand, smart grids have become an important development trend in the future power industry. Converter stations / substations, as a crucial component of the power Internet of Things (IoT) construction, face the problem of insufficient data sharing during their construction. This issue directly affects the stable operation and effective management of the power system. Rapidly achieving information synchronization between converter station / substation area networks and converged terminals is of great practical significance for accelerating the construction of smart grids.

[0003] Existing data sharing methods within converter stations / substations primarily rely on cloud-deployed business platforms to share various types of sensing data. However, this approach suffers from the following problems in practice: First, cloud computing employs a centralized network architecture, depending on stable network connections. Since the physical distance between the server and the converter station / substation is often considerable, data upload and download processes experience significant delays, reducing information sharing efficiency. Second, storing all converter station / substation data centrally on cloud servers poses a security risk; server failures may result in edge processing nodes being unable to access the data. Summary of the Invention

[0004] In view of this, the present invention provides a multi-source sensing data sharing method based on a consensus mechanism to solve the technical problems of low data transmission efficiency and insufficient security in the prior art.

[0005] In a first aspect, the present invention provides a multi-source sensing data sharing method based on a consensus mechanism, comprising:

[0006] Power nodes are classified using a clustering algorithm. Based on the classification results, several subnets are established. A proxy node is selected from the power nodes included in each subnet according to a proxy mechanism. The proxy nodes of each subnet are communicatively connected to the proxy nodes of other subnets, and the proxy nodes are also communicatively connected to other power nodes within the same subnet. Power nodes in the subnets are registered, and a public and private key is generated for each power node. When a power node needs to share data between subnets, it receives and broadcasts the sharing request from the power node acting as the data requester through a proxy node within the same subnet, and publishes the public key of the power node acting as the data requester. The proxy node also communicates with other power nodes within the same subnet. The proxy node in the subnet receives the sharing request and transmits the sharing request and the corresponding public key to the power node that is the data owner. Using the corresponding public key, the power node that is the data owner encrypts the data to be shared and uploads the encrypted data to the distributed database through the proxy node in the same subnet. The encrypted data is then transmitted through the distributed database to the power node that is the data requester, and the power node that is the data requester decrypts the encrypted data using its own private key. The proxy node in the same subnet as the power node that is the data requester broadcasts the transaction and stores the transaction record in the distributed database.

[0007] Optionally, classifying power nodes according to a clustering algorithm includes: dividing power nodes according to the K-Means algorithm, using the Euclidean distance between power nodes as a similarity and distance judgment criterion, and dividing the power nodes into several subnets with the objective of minimizing the sum of squared distances from all power nodes to the cluster center, wherein the formula representing the minimization of the objective is:

[0008]

[0009] In the formula, x i Let c represent each power node, and c represent the network to which the power node belongs. k and c j These are the k-th and j-th subnets, respectively. k For c k Cluster centers, ∥x i -μ k ∥ 2 This represents the sum of squared distances from power nodes to cluster centers.

[0010] Optionally, after selecting a proxy node from the power nodes included in each subnet according to the rights proxy mechanism, the method includes: using the proxy node as the root node, connecting other power nodes in the subnet using a tree network.

[0011] Optionally, after using the proxy node as the root node and connecting other power nodes in the subnet using a tree network, the method includes: when there is a data sharing requirement between power nodes in the same subnet, data transmission is completed through the connection of the tree network.

[0012] Optionally, after the data transmission is completed through the tree network connection, the process includes: broadcasting the data sharing behavior through proxy nodes within the same subnet, and storing the transaction record in the distributed database.

[0013] Optionally, the steps of generating a public key and a private key for each power node include: randomly generating a non-singular elliptic curve and selecting a base point located on the elliptic curve; selecting a random integer as the private key; and multiplying the base point and the private key to obtain the public key.

[0014] Optionally, before encrypting the data to be shared using the corresponding public key through the power node that owns the data, the process includes: using federated learning on an edge server to simplify the data to be shared, wherein, during the federated learning process, a stochastic gradient descent algorithm is used for training, and the optimization objective of the stochastic gradient descent algorithm is:

[0015]

[0016] Where: n is the number of instances of the data to be shared, arg w Here are the model parameters for the master node of the vertical federation model, and f(·) is the integration objective function running on the federation master node; y represents the primary knowledge computed by the k-th federated member for the j-th data point in its sub-model; m represents the number of members participating in the corresponding layer of federated learning; y represents the primary knowledge computed by the k-th federated member for the j-th data point in its sub-model. i Let be the i-th data label held by each federation member in the vertical federation model.

[0017] Optionally, the step of encrypting the data to be shared includes: encoding the data to be shared as a point on the elliptic curve; selecting a random number, and encrypting the encoded data to be shared based on the selected random number, the public key, and the base point to obtain a first ciphertext and a second ciphertext, wherein the calculation formulas for the first ciphertext and the second ciphertext are:

[0018] C1=R+g·Q

[0019] C2=g·G

[0020] Where R represents the encoded data to be shared, g is the selected random number, Q is the public key, G is the base point, and C1 and C2 represent the first ciphertext and the second ciphertext, respectively.

[0021] Optionally, the step of the power node acting as the data requester decrypting the encrypted data using its own private key includes: decrypting the first ciphertext and the second ciphertext using the power node's own private key to obtain the encoded data to be shared, wherein the decryption formula is:

[0022] R = C1 - S·C2

[0023] Where S is the private key; the encoded data to be shared is decoded to obtain the plaintext data.

[0024] Secondly, the present invention provides a multi-source sensing data sharing device based on a consensus mechanism, comprising:

[0025] The node classification module is used to classify power nodes according to a clustering algorithm, establish several subnets based on the classification results, and select proxy nodes from the power nodes included in each subnet according to the rights proxy mechanism. The proxy nodes of each subnet are communicatively connected to the proxy nodes of other subnets, and the proxy nodes are communicatively connected to other power nodes in the same subnet.

[0026] The node registration module is used to register power nodes in the subnet and generate a public key and a private key for each power node.

[0027] The announcement module is used to receive and broadcast the sharing request from the power node that is the data requester through the proxy node in the same subnet when the power node needs to share data between subnets, and to announce the public key of the power node that is the data requester.

[0028] The request receiving module is used to receive the sharing request through a proxy node that is in the same subnet as the power node that is the data owner, and to transmit the sharing request and the corresponding public key to the power node that is the data owner;

[0029] The data encryption module is used to encrypt the data to be shared by the power node that owns the data, using the corresponding public key, and then upload the encrypted data to the distributed database through the proxy node that is in the same subnet as the power node.

[0030] The data decryption module is used to transmit the encrypted data through the distributed database to the power node that is the data applicant, and the power node that is the data applicant decrypts the encrypted data according to its own private key.

[0031] The record storage module is used to broadcast the transaction through a proxy node that is in the same subnet as the power node that is the data requester, and to store the transaction record in the distributed database.

[0032] Thirdly, the present invention provides a computer device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the multi-source sensing data sharing method described in the first aspect or any corresponding embodiment thereof.

[0033] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the multi-source sensing data sharing method described in the first aspect or any corresponding embodiment thereof.

[0034] As can be seen from the above technical solutions, the embodiments of the present invention have the following advantages:

[0035] This invention provides a multi-source sensing data sharing method based on a consensus mechanism. It classifies power nodes using a clustering algorithm, establishes several subnets based on the classification results, and selects a proxy node from the power nodes included in each subnet according to a stake proxy mechanism. When power nodes need to share data between subnets, the proxy node in the same subnet receives and broadcasts the sharing request from the power node acting as the data requester, and publishes the public key of the power node acting as the data requester. The proxy node in the same subnet as the power node acting as the data owner receives the sharing request and transmits the sharing request and the corresponding public key to the power node acting as the data owner. Using the corresponding public key, the power node acting as the data owner encrypts the data to be shared and... The proxy nodes in the same subnet upload the encrypted data to the distributed database; the encrypted data is then transmitted through the distributed database to the power node acting as the data requester, which decrypts the encrypted data using its private key; the proxy nodes in the same subnet as the power node acting as the data requester broadcast the transaction and store the transaction record in the distributed database. This introduces a decentralized node consensus and distributed data storage scheme, which reduces the bandwidth requirements of the network compared to centralized transmission schemes, lowers the risk of storage failures and data leaks, greatly improves the efficiency of data sharing and the security of data transmission, and realizes the efficient sharing and storage of multi-source sensing data of power nodes within the power system.

[0036] Furthermore, based on the efficiency of distributed networks, the data transmission of power nodes is no longer subject to centralized data management systems, which supports the real-time requirements of power node wireless sensor networks, effectively reduces latency during transmission, and ensures the timeliness and reliability of power equipment data within power nodes. Attached Figure Description

[0037] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0038] Figure 1 This is a schematic diagram of the multi-source sensing data sharing network architecture according to an embodiment of the present invention;

[0039] Figure 2 This is a flowchart illustrating a multi-source sensing data sharing method according to an embodiment of the present invention;

[0040] Figure 3 This is a flowchart illustrating another multi-source sensing data sharing method according to an embodiment of the present invention;

[0041] Figure 4 This is a schematic diagram of the structure of a multi-source sensing data sharing device based on a consensus mechanism according to an embodiment of the present invention;

[0042] Figure 5 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0044] Converter stations / substations, as a crucial component of power transmission systems, are responsible for the transmission and conversion of electrical energy. They possess key functions such as converting transmission voltage and frequency, filtering current, and controlling current flow direction, directly impacting the stability and reliability of the power system. To ensure the safe and stable operation of the converter station / substation's area network, various sensors are required to acquire real-time operating data of the equipment. These include partial discharge sensors in DC converters, core current sensors for monitoring coupling core and clamp current, chemical monitoring sensors for monitoring ambient temperature and humidity, and inspection and monitoring equipment to ensure equipment safety. These monitoring devices connect to the fusion terminals in converter stations / substations, transmitting the collected data to the converter station / substation platform to facilitate continuous monitoring and in-depth analysis of the performance and operating environment of the converter station / substation equipment. On the other hand, to achieve unified power control, different converter stations / substations need to acquire real-time operating data from other converter stations / substations, such as transmission voltage, frequency, and current parameters, to achieve coordination, monitoring, and control of the overall converter station / substation area network. Furthermore, the complexity and scale of the power system require collaborative work and real-time data sharing among converter stations / substations to respond to various power system anomalies and emergencies. Therefore, there are significant data transmission and data sharing needs within the converter station / substation area network, posing a challenge to the data sharing efficiency of existing wireless sensor networks for converter stations / substations. To enhance the data sharing efficiency of wireless sensor networks in converter stations / substations and achieve high-speed transmission of data collected by various sensors and inspection equipment, this invention employs a clustering algorithm to divide the converter station / substation area network into multiple tree-shaped subnets. Data within the same subnet is transmitted using the network topology, while data from different subnets is connected and transmitted using a consensus mechanism. Furthermore, federated learning is used for data extraction, reducing data size and transmission latency.

[0045] This invention provides a consensus-based multi-source sensing data sharing application in a tree-structured network architecture. Specifically, this tree-structured network architecture includes several subnets. Each subnet categorizes power nodes using a clustering algorithm, identifying power nodes belonging to their respective subnets based on the classification results. Each subnet selects a proxy node from its included power nodes according to a staked proxy mechanism. The proxy nodes of each subnet are communicatively connected to proxy nodes in other subnets, and the proxy nodes are also communicatively connected to other power nodes within the same subnet. After establishing a subnet, the power nodes within it are registered, generating public and private keys for each node. When power nodes need to exchange data between subnets... During sharing, a proxy node in the same subnet receives and broadcasts a sharing request from a power node acting as the data requester, and publishes the public key of the power node acting as the data requester. A proxy node in the same subnet as the power node acting as the data owner receives the sharing request and transmits the sharing request and the corresponding public key to the power node acting as the data owner. Using the corresponding public key, the power node acting as the data owner encrypts the data to be shared and uploads the encrypted data to a distributed database through a proxy node in the same subnet. The encrypted data is then transmitted through the distributed database to the power node acting as the data requester, and the power node acting as the data requester decrypts the encrypted data using its own private key.

[0046] Specifically, the power node is a converter station / substation, for example, such as Figure 1 As shown, the power node is the converter station node.

[0047] A consensus-based multi-source sensing data sharing network architecture, such as... Figure 1As shown, converter stations / substations are connected in a tree-like network. Based on their specific topological location, the entire converter station / substation area network can be divided into multiple different tree-like subnets, each with a defined affiliation. Within a subnet, converter station / substation nodes use a Delegated Proof of Stake (DPOS) consensus mechanism to select a proxy node, which is responsible for connecting other subnets. Thus, if converter station / substation nodes within the same subnet need to share data, they can transmit the data through the tree-like network connection, while the proxy node broadcasts the data across the network, recording the data sharing activity in a distributed database. If converter station / substation nodes in different tree-like networks need to share data, the data requester submits a request to the network through the proxy node. After the request is accepted, the data owner uses federated learning to reduce the data size, encrypts the data using a public key, and entrusts the proxy node to upload it to the distributed database. The data requester retrieves the data using its private key, and the proxy node broadcasts the data across the network, recording the data sharing activity in the database.

[0048] The consensus-based multi-source sensing data sharing network architecture of this invention has the following advantages:

[0049] First, a distributed network connection is adopted within the converter station / substation domain. Network nodes can be composed of edge servers, which are located at the network edge, physically close to the converter station / substation, effectively reducing information sharing latency. Second, a consensus mechanism is adopted, using encryption algorithms, smart contracts, and other technologies to achieve decentralized point-to-point transactions within the converter station / substation domain. Information sharing records between nodes are encrypted and stored as underlying information in a distributed database within the network, ensuring the security of shared data and preventing data loss and leakage. Third, a DPOS mechanism is adopted, selecting proxy nodes within the subnet to handle data sharing between different subnets. Fourth, a tree-structured subnet network architecture is used, reducing the number of data transfers within the converter station / substation domain network and lowering transmission latency.

[0050] This invention proposes a multi-source sensing data sharing method based on a consensus mechanism, which is applied to the network architecture described in the above embodiments, such as... Figure 2 and Figure 3 As shown, the method includes:

[0051] Step S201: Classify the power nodes according to the clustering algorithm, establish several subnets based on the classification results, and select a proxy node from the power nodes included in each subnet according to the rights proxy mechanism. The proxy node of each subnet is communicatively connected to the proxy nodes of other subnets, and the proxy node is communicatively connected to other power nodes in the same subnet.

[0052] Specifically, in this embodiment of the invention, the power node is a power node.

[0053] Initialize the power node devices, distributed network configuration, and edge server configuration for federated learning. Use the K-Means clustering algorithm to partition the power node network, selecting Euclidean distance between power nodes as the similarity and distance criterion. Divide the network into K subnets. The goal is to minimize the sum of squared distances from all power nodes to the cluster centers, thus achieving the segmentation of the power node network. The formula for minimizing the objective is:

[0054]

[0055] In the formula, x i Let c represent each power node, and c represent the network to which the power node belongs. k and c j These are the k-th and j-th subnets, respectively. k For c k Cluster centers, ||x i -μ k || 2 This represents the sum of squared distances from the power nodes to the cluster centers. After the network is partitioned, each subnet selects a proxy node based on the DPOS consensus mechanism. Using the proxy node as the root, a tree-like network topology connects other nodes within the subnet. A distributed network architecture is constructed by partitioning the network into subnets.

[0056] Step S202: Register the power nodes in the subnet and generate a public key and a private key for each power node.

[0057] Specifically, power nodes register in the network and generate a public key P and a private key S based on an elliptic curve cryptography algorithm.

[0058] Select an elliptic curve E:y 2 =x 3 +αx+b, where a and b are curve parameters, satisfying 4a 3 +27b 2 ≠0, to ensure that the curve is non-singular.

[0059] Choose a base point G on the elliptic curve E, with order n, and have nG = θ, where θ represents the point at infinity and n is a prime number.

[0060] Choose a random integer S as the private key, satisfying 1≤S≤n, and calculate the public key Q according to the formula Q=S·G.

[0061] Step S203: When a power node needs to share data between subnets, it receives and broadcasts the sharing request from the power node that is the data requester through the proxy node in the same subnet, and publishes the public key of the power node that is the data requester.

[0062] Specifically, the power node, acting as the data applicant, entrusts its proxy node to initiate and broadcast a sharing request between subnets, and submits the power node's public key Q as an identifier.

[0063] Step S204: Receive the sharing request through the proxy node that is in the same subnet as the power node that is the data owner, and transmit the sharing request and the corresponding public key to the power node that is the data owner.

[0064] Specifically, the proxy node of the power node, which is the data owner, accepts the sharing request and transmits the sharing request and public key Q to the power node, which is the data owner, through the topology network within the subnet.

[0065] Step S205: Using the corresponding public key, the data to be shared is encrypted by the power node that owns the data, and then uploaded to the distributed database by the proxy node that is in the same subnet as the power node.

[0066] Upon receiving a sharing request and public key Q, the power node, acting as the data owner, retrieves the data to be shared based on the request, encrypts the data, and uploads the encrypted data to the distributed database through a proxy node on the same subnet. Specifically, the encryption process is as follows:

[0067] The data to be shared is encoded as a point R on the elliptic curve E, which is the encoded data to be shared;

[0068] Choose a random number g, and encrypt the encoded data R to be shared based on the chosen random number g, the public key Q, and the base point G to obtain the first ciphertext and the second ciphertext. The formulas for calculating the first ciphertext and the second ciphertext are as follows:

[0069] C1=R+g·Q

[0070] C2=g·G

[0071] Where R represents the encoded data to be shared, g is the chosen random number, Q is the public key, G is the base point, and C1 and C2 represent the first ciphertext and the second ciphertext, respectively. The first ciphertext and the second ciphertext are the ciphertext data.

[0072] Step S206: The encrypted data is transmitted to the power node that is the data applicant through the distributed database, and the power node that is the data applicant decrypts the encrypted data according to its own private key.

[0073] Specifically, the corresponding decryption process is as follows:

[0074] The first and second ciphertexts are decrypted using the private key of the power node that is the data requester, resulting in the encoded data to be shared. The decryption formula is as follows:

[0075] R = C1 - S·C2

[0076] Where S is the private key. The encoded data to be shared is decoded to obtain the plaintext data.

[0077] Step S207: Broadcast the transaction through a proxy node that is in the same subnet as the power node that is the data requester, and store the transaction record in the distributed database.

[0078] After the data sharing parties complete the data transmission, the agent node of the data applicant broadcasts the transaction and stores the smart contract record of the transaction in the distributed database.

[0079] An embodiment of the present invention provides a multi-source sensing data sharing method based on a consensus mechanism.

[0080] By employing a clustering-based subnetting mechanism and DPOS consensus technology, the low data transmission efficiency and insufficient security of existing centralized transmission schemes in converter station / substation area networks were successfully overcome. By introducing decentralized node consensus and a distributed data storage scheme, compared to centralized transmission schemes, the bandwidth requirements of the network are reduced, the risk of storage failures and data leaks is lowered, and the efficiency of data sharing and the security of data transmission are greatly improved. This enables efficient sharing and storage of multi-source sensing data in power system converter station / substation area networks.

[0081] The stability and real-time performance of converter station / substation area networks have been effectively improved. Based on the efficiency of distributed networks, data transmission in converter stations / substations is no longer constrained by centralized data management systems, supporting the real-time requirements of wireless sensor networks in converter stations / substations. This effectively reduces latency during transmission, ensuring the timeliness and reliability of power equipment data within the converter station / substation. Simultaneously, based on smart contracts and consensus mechanisms, converter station / substation area networks can achieve refined data sharing and access control, resolving data privacy and security issues, reducing the risks of data forgery and lack of trust, and providing strong support for the stable operation of converter stations / substations.

[0082] High data transmission efficiency. Leveraging a distributed network architecture, data transmission between converter stations / substations can be directly completed through the distributed network, reducing latency during transmission. Converter stations / substations can use proxy nodes within the subnet to achieve rapid data relay within the network, improving data transmission efficiency and speed. For example, video data collected by unmanned inspection equipment in converter stations / substations can be quickly transmitted using this method to complete monitoring tasks. This means that monitoring personnel can quickly obtain real-time video streams, rapidly identify potential problems, and take necessary measures, thereby ensuring the safe operation and maintenance of power equipment in converter stations / substations and improving the reliability and security of the converter station / substation system.

[0083] Enhanced data trustworthiness. Through a decentralized consensus mechanism and distributed verification methods, the authenticity and credibility of data from converter stations / substations are guaranteed. Within the network, every piece of data is recorded in an immutable distributed database, and each record contains information from the previous one, forming a chain structure. Once data is recorded in the database, it cannot be tampered with or deleted, ensuring data integrity and reducing data disputes and trust issues.

[0084] Flexible data sharing and access control. A consensus mechanism enables data sharing and access control between converter stations / substations, allowing for flexible setting of different permissions and roles for different converter stations / substations, achieving fine-grained control over shared data. Converter stations / substations can authorize and verify permissions via the network, ensuring legitimate access and use of data while also protecting data privacy and security.

[0085] The embodiments of the present invention greatly improve the efficiency of sharing sensing data and the security of data transmission based on the power wireless sensor network system. It enables the efficient sharing and storage of multi-source sensing data by the edge computing communication nodes of the power wireless sensor network in scenarios such as substations and converter stations, supporting edge applications for status analysis and fault diagnosis of power equipment.

[0086] In an optional implementation, after using the proxy node as the root node and connecting other power nodes within the subnet using a tree network in step S201, the following steps are included:

[0087] When power nodes within the same subnet have a need for data sharing, data transmission is accomplished through tree-structured network connections.

[0088] Tree-structured networks help establish stable data transmission paths between converter station / substation nodes, improving the efficiency and reliability of data sharing.

[0089] Furthermore, after the data transmission is completed through the tree network connection, the process includes: broadcasting the data sharing behavior through proxy nodes within the same subnet, and storing the transaction record in a distributed database.

[0090] Specifically, if power nodes within the same subnet have a need for data sharing, they can complete the data transmission through a tree-structured network connection. At the same time, the data sharing behavior is broadcast in the network by a proxy node and recorded in a distributed database.

[0091] By broadcasting this data sharing activity and storing the transaction record in a distributed database, decentralized node consensus and distributed data storage are achieved, enabling efficient sharing and storage of multi-source sensing data in the power system converter station / substation area network.

[0092] In an optional implementation, before encrypting the data to be shared by the power node that is the data owner, using the corresponding public key in step S205, the process includes: using federated learning on an edge server to simplify the data to be shared, wherein, during the federated learning process, a stochastic gradient descent algorithm is used for training, and the optimization objective of the stochastic gradient descent algorithm is:

[0093]

[0094] Where: n is the number of instances of the data to be shared, arg w Here are the model parameters for the master node of the vertical federation model, and f(·) is the integration objective function running on the federation master node; y represents the primary knowledge computed by the k-th federated member for the j-th data point in its sub-model; m represents the number of members participating in the corresponding layer of federated learning; y represents the primary knowledge computed by the k-th federated member for the j-th data point in its sub-model. i Let be the i-th data label held by each federation member in the vertical federation model.

[0095] By adopting federated learning technology, the amount of data transmission between converter stations / substations can be reduced while protecting data privacy, thereby improving the efficiency and security of data sharing.

[0096] A multi-source sensing data sharing method based on a consensus mechanism according to an embodiment of the present invention has the following effects:

[0097] By recording data sharing behavior between converter stations and substations using a distributed database, decentralized, trustworthy, and tamper-proof data sharing is achieved. This consensus-based data sharing method offers high security and effectively meets the information sharing needs between converter stations and substations.

[0098] This study effectively utilizes a tree-structured converter station / substation network. A clustering algorithm divides the network into multiple distinct tree-like subnets, each containing multiple converter station / substation nodes with a defined affiliation, thus achieving efficient connection and organization of the network. This tree-like topology facilitates the establishment of stable data transmission paths between converter station / substation nodes, improving the efficiency and reliability of data sharing.

[0099] Application of DPOS consensus mechanism. By adopting the DPOS consensus mechanism, proxy nodes in the subnet are selected to be responsible for communicating with the network and recording data transmission, which ensures the high-speed and stable operation of the network and improves the efficiency of data sharing between converter stations / substations.

[0100] Application of federated learning technology. By adopting federated learning technology, the amount of data transmission between converter stations / substations can be reduced, improving the efficiency and security of data sharing while protecting data privacy.

[0101] This invention also provides a multi-source sensing data sharing device based on a consensus mechanism, such as... Figure 4 As shown, it includes:

[0102] The node classification module is used to classify power nodes according to a clustering algorithm, establish several subnets based on the classification results, and select proxy nodes from the power nodes included in each subnet according to the rights proxy mechanism. The proxy nodes of each subnet are communicatively connected to the proxy nodes of other subnets, and the proxy nodes are communicatively connected to other power nodes in the same subnet.

[0103] The node registration module is used to register power nodes in the subnet and generate a public key and a private key for each power node.

[0104] The announcement module is used to receive and broadcast the sharing request from the power node that is the data requester through the proxy node in the same subnet when the power node needs to share data between subnets, and to announce the public key of the power node that is the data requester.

[0105] The request receiving module is used to receive the sharing request through a proxy node that is in the same subnet as the power node that is the data owner, and to transmit the sharing request and the corresponding public key to the power node that is the data owner;

[0106] The data encryption module is used to encrypt the data to be shared by the power node that owns the data, using the corresponding public key, and then upload the encrypted data to the distributed database through the proxy node that is in the same subnet as the power node.

[0107] The data decryption module is used to transmit the encrypted data through the distributed database to the power node that is the data applicant, and the power node that is the data applicant decrypts the encrypted data according to its own private key.

[0108] The record storage module is used to broadcast the transaction through a proxy node that is in the same subnet as the power node that is the data requester, and to store the transaction record in the distributed database.

[0109] This invention also provides a schematic diagram of the structure of a computer device, such as... Figure 5 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 5 Take a processor 10 as an example.

[0110] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0111] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0112] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0113] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0114] The computer device also includes an input device 30 and an output device 40. The processor 10, memory 20, input device 30, and output device 40 can be connected via a bus or other means. Figure 5 Taking the example of a connection between China and Israel via a bus.

[0115] Input device 30 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the computer device, such as a touchscreen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 40 may include display devices, auxiliary lighting devices (e.g., LEDs), and haptic feedback devices (e.g., vibration motors). The aforementioned display devices include, but are not limited to, liquid crystal displays, light-emitting diodes, displays, and plasma displays. In some alternative embodiments, the display device may be a touchscreen.

[0116] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0117] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A multi-source sensing data sharing method based on a consensus mechanism, characterized in that, include: Power nodes are classified according to a clustering algorithm. Based on the classification results, several subnets are established. In accordance with the rights proxy mechanism, proxy nodes are selected from the power nodes included in each subnet. The proxy nodes of each subnet are communicatively connected to the proxy nodes of other subnets, and the proxy nodes are communicatively connected to other power nodes in the same subnet. Register the power nodes in the subnet and generate a public key and a private key for each power node; When a power node needs to share data between subnets, it receives the sharing request from the power node that is the data requester through a proxy node in the same subnet and broadcasts it, and publishes the public key of the power node that is the data requester. The proxy node, which is in the same subnet as the power node that owns the data, receives the sharing request and transmits the sharing request and the corresponding public key to the power node that owns the data. Using the corresponding public key, the power node, which is the data owner, encrypts the data to be shared and uploads the encrypted data to the distributed database through a proxy node in the same subnet. The encrypted data is transmitted through the distributed database to the power node that is the data applicant, and the power node decrypts the encrypted data using its own private key. The transaction is broadcast by a proxy node that is in the same subnet as the power node that is the data requester, and the transaction record is stored in the distributed database. The steps for generating the public and private keys for each power node include: Randomly generate a non-singular elliptic curve and select a base point located on the elliptic curve; Choose a random integer as the private key; Multiply the base point and the private key to obtain the public key; This includes, before encrypting the data to be shared using the corresponding public key through the power node that owns the data, the following: Federated learning is used on edge servers to streamline the shared data. During the federated learning process, stochastic gradient descent is used for training, and the optimization objective of stochastic gradient descent is: in: The number of instances of the data to be shared. These are the master node model parameters for the vertical federation model. To integrate the objective function running on the federated master node; For the first The federal members targeted their sub-models for the first... Basic knowledge derived from data calculations; The number of members participating in the corresponding layer of federated learning; The number of federal members in a vertical federal model Data tags.

2. The multi-source sensing data sharing method according to claim 1, characterized in that, The classification of power nodes based on clustering algorithms includes: The power nodes are partitioned using the K-Means algorithm, with the Euclidean distance between power nodes used as the similarity and distance criterion. The goal is to minimize the sum of squared distances from all power nodes to the cluster centers, thus dividing the power nodes into several subnets. The formula representing the minimization of the objective is: In the formula, For each power node, The network to which the power node belongs. and The first division is respectively the second division and the Individual Network for Cluster centers This represents the sum of squared distances from power nodes to cluster centers.

3. The multi-source sensing data sharing method according to claim 1, characterized in that, After selecting a proxy node from among the power nodes included in each subnet according to the rights proxy mechanism, the process includes: Using the proxy node as the root node, a tree network is used to connect other power nodes within the subnet.

4. The multi-source sensing data sharing method according to claim 3, characterized in that, After using the proxy node as the root node and connecting other power nodes within the subnet using a tree network, the following is included: When power nodes within the same subnet have a need for data sharing, data transmission is accomplished through tree-structured network connections.

5. The multi-source sensing data sharing method according to claim 4, characterized in that, After data transmission is completed through a tree-structured network connection, the following steps are included: The data sharing activity is broadcast by proxy nodes within the same subnet, and the transaction record is stored in the distributed database.

6. The multi-source sensing data sharing method according to claim 1, characterized in that, The steps involved in encrypting shared data include: The data to be shared is encoded as a point on the elliptic curve; Select a random number, and encrypt the encoded data to be shared based on the selected random number, the public key, and the base point to obtain a first ciphertext and a second ciphertext. The calculation formulas for the first ciphertext and the second ciphertext are as follows: in, This represents the encoded data to be shared. For the selected random number, For the public key, The base point, and These represent the first ciphertext and the second ciphertext, respectively.

7. The multi-source sensing data sharing method according to claim 6, characterized in that, The steps for a power node, as the data requester, to decrypt encrypted data using its own private key include: The first and second ciphertexts are decrypted using the private key of the power node that is the data requester, resulting in the encoded data to be shared. The decryption formula is as follows: in, For private key; Decode the encoded data to be shared to obtain plaintext data.

8. A multi-source sensing data sharing device based on a consensus mechanism, characterized in that, include: The node classification module is used to classify power nodes according to a clustering algorithm, establish several subnets based on the classification results, and select proxy nodes from the power nodes included in each subnet according to the rights proxy mechanism. The proxy nodes of each subnet are communicatively connected to the proxy nodes of other subnets, and the proxy nodes are communicatively connected to other power nodes in the same subnet. The node registration module is used to register power nodes in the subnet and generate a public key and a private key for each power node. The announcement module is used to receive and broadcast the sharing request from the power node that is the data requester through the proxy node in the same subnet when the power node needs to share data between subnets, and to announce the public key of the power node that is the data requester. The request receiving module is used to receive the sharing request through a proxy node that is in the same subnet as the power node that is the data owner, and to transmit the sharing request and the corresponding public key to the power node that is the data owner; The data encryption module is used to encrypt the data to be shared by the power node that owns the data, using the corresponding public key, and then upload the encrypted data to the distributed database through the proxy node that is in the same subnet as the power node. The data decryption module is used to transmit the encrypted data through the distributed database to the power node that is the data applicant, and the power node that is the data applicant decrypts the encrypted data according to its own private key. The record storage module is used to broadcast the transaction through a proxy node that is in the same subnet as the power node that is the data requester, and to store the transaction record in the distributed database; The steps for generating the public and private keys for each power node include: Randomly generate a non-singular elliptic curve and select a base point located on the elliptic curve; Choose a random integer as the private key; Multiply the base point and the private key to obtain the public key; This includes, before encrypting the data to be shared using the corresponding public key through the power node that owns the data, the following: Federated learning is used on edge servers to streamline the shared data. During the federated learning process, stochastic gradient descent is used for training, and the optimization objective of stochastic gradient descent is: in: The number of instances of the data to be shared. These are the master node model parameters for the vertical federation model. To integrate the objective function running on the federated master node; For the first The federal members targeted their sub-models for the first... Basic knowledge derived from data calculations; The number of members participating in the corresponding layer of federated learning; The number of federal members in a vertical federal model Data tags.

9. A computer device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the computer instructions to perform the multi-source sensing data sharing method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the multi-source sensing data sharing method according to any one of claims 1 to 7.

Images